2 Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
5 Redistribution and use in source and binary forms, with or without
6 modification, are permitted provided that the following conditions
9 * Redistributions of source code must retain the above copyright
10 notice, this list of conditions and the following disclaimer.
11 * Redistributions in binary form must reproduce the above copyright
12 notice, this list of conditions and the following disclaimer in
13 the documentation and/or other materials provided with the
15 * Neither the name of Intel Corporation nor the names of its
16 contributors may be used to endorse or promote products derived
17 from this software without specific prior written permission.
19 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20 "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21 LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22 A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23 OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24 SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25 LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26 DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27 THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29 OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
31 QoS Scheduler Sample Application
32 ================================
34 The QoS sample application demonstrates the use of the DPDK to provide QoS scheduling.
39 The architecture of the QoS scheduler application is shown in the following figure.
41 .. _figure_qos_sched_app_arch:
43 .. figure:: img/qos_sched_app_arch.*
45 QoS Scheduler Application Architecture
48 There are two flavors of the runtime execution for this application,
49 with two or three threads per each packet flow configuration being used.
50 The RX thread reads packets from the RX port,
51 classifies the packets based on the double VLAN (outer and inner) and
52 the lower two bytes of the IP destination address and puts them into the ring queue.
53 The worker thread dequeues the packets from the ring and calls the QoS scheduler enqueue/dequeue functions.
54 If a separate TX core is used, these are sent to the TX ring.
55 Otherwise, they are sent directly to the TX port.
56 The TX thread, if present, reads from the TX ring and write the packets to the TX port.
58 Compiling the Application
59 -------------------------
61 To compile the sample application see :doc:`compiling`.
63 The application is located in the ``qos_sched`` sub-directory.
67 This application is intended as a linuxapp only.
71 To get statistics on the sample app using the command line interface as described in the next section,
72 DPDK must be compiled defining *CONFIG_RTE_SCHED_COLLECT_STATS*,
73 which can be done by changing the configuration file for the specific target to be compiled.
75 Running the Application
76 -----------------------
80 In order to run the application, a total of at least 4
81 G of huge pages must be set up for each of the used sockets (depending on the cores in use).
83 The application has a number of command line options:
85 .. code-block:: console
87 ./qos_sched [EAL options] -- <APP PARAMS>
89 Mandatory application parameters include:
91 * --pfc "RX PORT, TX PORT, RX LCORE, WT LCORE, TX CORE": Packet flow configuration.
92 Multiple pfc entities can be configured in the command line,
93 having 4 or 5 items (if TX core defined or not).
95 Optional application parameters include:
97 * -i: It makes the application to start in the interactive mode.
98 In this mode, the application shows a command line that can be used for obtaining statistics while
99 scheduling is taking place (see interactive mode below for more information).
101 * --mst n: Master core index (the default value is 1).
103 * --rsz "A, B, C": Ring sizes:
105 * A = Size (in number of buffer descriptors) of each of the NIC RX rings read
106 by the I/O RX lcores (the default value is 128).
108 * B = Size (in number of elements) of each of the software rings used
109 by the I/O RX lcores to send packets to worker lcores (the default value is 8192).
111 * C = Size (in number of buffer descriptors) of each of the NIC TX rings written
112 by worker lcores (the default value is 256)
114 * --bsz "A, B, C, D": Burst sizes
116 * A = I/O RX lcore read burst size from the NIC RX (the default value is 64)
118 * B = I/O RX lcore write burst size to the output software rings,
119 worker lcore read burst size from input software rings,QoS enqueue size (the default value is 64)
121 * C = QoS dequeue size (the default value is 32)
123 * D = Worker lcore write burst size to the NIC TX (the default value is 64)
125 * --msz M: Mempool size (in number of mbufs) for each pfc (default 2097152)
127 * --rth "A, B, C": The RX queue threshold parameters
129 * A = RX prefetch threshold (the default value is 8)
131 * B = RX host threshold (the default value is 8)
133 * C = RX write-back threshold (the default value is 4)
135 * --tth "A, B, C": TX queue threshold parameters
137 * A = TX prefetch threshold (the default value is 36)
139 * B = TX host threshold (the default value is 0)
141 * C = TX write-back threshold (the default value is 0)
143 * --cfg FILE: Profile configuration to load
145 Refer to *DPDK Getting Started Guide* for general information on running applications and
146 the Environment Abstraction Layer (EAL) options.
148 The profile configuration file defines all the port/subport/pipe/traffic class/queue parameters
149 needed for the QoS scheduler configuration.
151 The profile file has the following format:
155 ; port configuration [port]
158 number of subports per port = 1
159 number of pipes per subport = 4096
160 queue sizes = 64 64 64 64
162 ; Subport configuration
165 tb rate = 1250000000; Bytes per second
166 tb size = 1000000; Bytes
167 tc 0 rate = 1250000000; Bytes per second
168 tc 1 rate = 1250000000; Bytes per second
169 tc 2 rate = 1250000000; Bytes per second
170 tc 3 rate = 1250000000; Bytes per second
171 tc period = 10; Milliseconds
172 tc oversubscription period = 10; Milliseconds
174 pipe 0-4095 = 0; These pipes are configured with pipe profile 0
179 tb rate = 305175; Bytes per second
180 tb size = 1000000; Bytes
182 tc 0 rate = 305175; Bytes per second
183 tc 1 rate = 305175; Bytes per second
184 tc 2 rate = 305175; Bytes per second
185 tc 3 rate = 305175; Bytes per second
186 tc period = 40; Milliseconds
188 tc 0 oversubscription weight = 1
189 tc 1 oversubscription weight = 1
190 tc 2 oversubscription weight = 1
191 tc 3 oversubscription weight = 1
193 tc 0 wrr weights = 1 1 1 1
194 tc 1 wrr weights = 1 1 1 1
195 tc 2 wrr weights = 1 1 1 1
196 tc 3 wrr weights = 1 1 1 1
198 ; RED params per traffic class and color (Green / Yellow / Red)
201 tc 0 wred min = 48 40 32
202 tc 0 wred max = 64 64 64
203 tc 0 wred inv prob = 10 10 10
204 tc 0 wred weight = 9 9 9
206 tc 1 wred min = 48 40 32
207 tc 1 wred max = 64 64 64
208 tc 1 wred inv prob = 10 10 10
209 tc 1 wred weight = 9 9 9
211 tc 2 wred min = 48 40 32
212 tc 2 wred max = 64 64 64
213 tc 2 wred inv prob = 10 10 10
214 tc 2 wred weight = 9 9 9
216 tc 3 wred min = 48 40 32
217 tc 3 wred max = 64 64 64
218 tc 3 wred inv prob = 10 10 10
219 tc 3 wred weight = 9 9 9
224 These are the commands that are currently working under the command line interface:
228 * --quit: Quits the application.
232 * stats app: Shows a table with in-app calculated statistics.
234 * stats port X subport Y: For a specific subport, it shows the number of packets that
235 went through the scheduler properly and the number of packets that were dropped.
236 The same information is shown in bytes.
237 The information is displayed in a table separating it in different traffic classes.
239 * stats port X subport Y pipe Z: For a specific pipe, it shows the number of packets that
240 went through the scheduler properly and the number of packets that were dropped.
241 The same information is shown in bytes.
242 This information is displayed in a table separating it in individual queues.
246 All of these commands work the same way, averaging the number of packets throughout a specific subset of queues.
248 Two parameters can be configured for this prior to calling any of these commands:
250 * qavg n X: n is the number of times that the calculation will take place.
251 Bigger numbers provide higher accuracy. The default value is 10.
253 * qavg period X: period is the number of microseconds that will be allowed between each calculation.
254 The default value is 100.
256 The commands that can be used for measuring average queue size are:
258 * qavg port X subport Y: Show average queue size per subport.
260 * qavg port X subport Y tc Z: Show average queue size per subport for a specific traffic class.
262 * qavg port X subport Y pipe Z: Show average queue size per pipe.
264 * qavg port X subport Y pipe Z tc A: Show average queue size per pipe for a specific traffic class.
266 * qavg port X subport Y pipe Z tc A q B: Show average queue size of a specific queue.
271 The following is an example command with a single packet flow configuration:
273 .. code-block:: console
275 ./qos_sched -l 1,5,7 -n 4 -- --pfc "3,2,5,7" --cfg ./profile.cfg
277 This example uses a single packet flow configuration which creates one RX thread on lcore 5 reading
278 from port 3 and a worker thread on lcore 7 writing to port 2.
280 Another example with 2 packet flow configurations using different ports but sharing the same core for QoS scheduler is given below:
282 .. code-block:: console
284 ./qos_sched -l 1,2,6,7 -n 4 -- --pfc "3,2,2,6,7" --pfc "1,0,2,6,7" --cfg ./profile.cfg
286 Note that independent cores for the packet flow configurations for each of the RX, WT and TX thread are also supported,
287 providing flexibility to balance the work.
289 The EAL coremask/corelist is constrained to contain the default mastercore 1 and the RX, WT and TX cores only.
294 The Port/Subport/Pipe/Traffic Class/Queue are the hierarchical entities in a typical QoS application:
296 * A subport represents a predefined group of users.
298 * A pipe represents an individual user/subscriber.
300 * A traffic class is the representation of a different traffic type with a specific loss rate,
301 delay and jitter requirements; such as data voice, video or data transfers.
303 * A queue hosts packets from one or multiple connections of the same type belonging to the same user.
305 The traffic flows that need to be configured are application dependent.
306 This application classifies based on the QinQ double VLAN tags and the IP destination address as indicated in the following table.
308 .. _table_qos_scheduler_1:
310 .. table:: Entity Types
312 +----------------+-------------------------+--------------------------------------------------+----------------------------------+
313 | **Level Name** | **Siblings per Parent** | **QoS Functional Description** | **Selected By** |
315 +================+=========================+==================================================+==================================+
316 | Port | - | Ethernet port | Physical port |
318 +----------------+-------------------------+--------------------------------------------------+----------------------------------+
319 | Subport | Config (8) | Traffic shaped (token bucket) | Outer VLAN tag |
321 +----------------+-------------------------+--------------------------------------------------+----------------------------------+
322 | Pipe | Config (4k) | Traffic shaped (token bucket) | Inner VLAN tag |
324 +----------------+-------------------------+--------------------------------------------------+----------------------------------+
325 | Traffic Class | 4 | TCs of the same pipe services in strict priority | Destination IP address (0.0.X.0) |
327 +----------------+-------------------------+--------------------------------------------------+----------------------------------+
328 | Queue | 4 | Queue of the same TC serviced in WRR | Destination IP address (0.0.0.X) |
330 +----------------+-------------------------+--------------------------------------------------+----------------------------------+
332 Please refer to the "QoS Scheduler" chapter in the *DPDK Programmer's Guide* for more information about these parameters.